Tool Box Newsletter Logo

 A computer newsletter for translation professionals


Issue 12-07-211
(the two hundred eleventh edition)  
Contents
1. Service Packing
2. Verifika (Premium Edition)
3. MTV -- Machine Translation Views
4. Excelling Glossary Searches
5. The Panic-Stricken Translator (Premium Edition)
6. New Password for the Tool Box Newsletter Archive
The Last Word on the Tool Kit
Perfect!

I just rediscovered the poetry and writing of Sherman Alexie, a Coeur d'Alene/Spokane Indian. Here is a little blurb from his collection One Stick Song that made me laugh out loud as I was driving through Spokane, Washington, last week:

"Sherman," says the critic, "How does the oral tradition apply to your work?"

"Well," I say, as I hold my latest book close to me, "It really doesn't apply at all because I typed this. And when I'm typing, I'm really, really quiet."

Perfect!

Speaking of (perfect) books, Nataly Kelly's and my love letter to translators and interpreters, Found in Translation, will soon be rolling off the presses and into the hands of all who love language. We're thrilled with the advance praise the book is receiving and we know that you and your friends and clients will enjoy it, too.

Here is another perfect thought -- or maybe I should call it a prfct thght. Rina Ne'eman shared this article recently about how texting is causing a Hebrewization of English. To save space, many texters are simply eliminating vowels in texting, and yet the message remains completely understandable (not to me, mind you, but my kids certainly seem to understand it). And finally I realized why it's pssbl 2 nderstnd clsscl vwl-lss Hbrw.

This fit so perfectly with a question I was recently asked in an interview. The rather pessimistic interviewer asked for my opinion on the future of translation, given the fact that kids no longer write. Man, I was all over him for that question! Kids don't write?! They write so much more than most of us (assuming that most adults, like me, still text less than we talk). My teenagers easily send more than 150 texts a day, and I did not write that much in a week as a 15-year-old! I believe we can safely say that our profession will be in capable hands (and if you doubt the validity of texting as "real" writing, check out this TED Talent Search talk by John McWhorter).

1. Service Packing

In the distant past, Trados made it a habit to really embarrass itself with full, new (and paid) releases that would have been a sub-point release for some of its competitors. For the last few versions of Trados, however, SDL has come out with meaningful Service Pack releases that actually add interesting functionality to the existing version of Trados... and are free.

This is also true for the upcoming release (which should be out sometime this week or so) of Trados 2011 SP2.

I had a chance to chat with a few folks from SDL about it, including Massi Ghislandi and Daniel Brockmann, and they showed me what's new. The main new feature is clever enough that I think we will quickly see it in competing tools as well.

Of course, there is a host of usual enhancements that I won't list in exhaustive detail (such as faster processing, better support of right-to-left languages, support for new versions of InDesign and DITA, the ability to edit the source document, increased customizability of AutoSuggest, etc. -- in short, all the stuff that would have warranted a full release just a few years ago), but there is also at least one feature that is truly new.

This is the ability to process track-changed Word documents. While this might not sound too impressive, it represents far more than a simple fix of Trados' previous frustrating inability to process track-changed documents (every time I asked myself: Why can't they simply accept the changes and inform me of that?). It also includes the ability to leave the tracked changes (those struck-out and underlined passages in a Word document that show you what edits have been made to it) and actually display them in the source pane in Trados in exactly the same way you see them in Word. This allows you to create a translated document with incorporated tracked changes -- if that's what you want. Or you can just leave the tracked changes in the source document for information's sake, or hide them, or accept them during the import of the documents.

SDL's main reason for introducing this capability is that it's an extremely relevant feature for some of its core industries, including pharmaceutical and medical -- those industries that require strict regulatory processes, including the necessity to track back the genesis of documents. I've heard folks complain about SDL's emphasis on those industries, but this particular feature is useful for many others as well.

The TM matching when encountering formerly perfect matches of now track-changed segments marks those matches with a special "TC" status and considers them perfect matches for the sake of the PerfectMatch feature. (If this last sentence didn't make any sense to you, don't worry -- it will make sense to most existing Trados users.)

A number of other TEnTs allow you to output a translated version of documents into a bilingual Word document where they can either be edited or translated from scratch by someone who does not own the originating tool. The resulting translated document can then be reimported back into the translation environment tool and everything is hunky-dory. Trados now also offers this feature in its upcoming service pack, but what's really interesting is how SDL decided to incorporate it.

For the last couple of years, SDL has been running the OpenExchange program. Styled after Apple's App Store, it invited developers to come up with apps that users could use in combination with Trados. I think it's fair to say that it did not take off as quickly as SDL had hoped, but there are some interesting apps available, some paid and some for free. One great benefit of the program for SDL is that it provides really good insight into what users want. If an app is very popular, it's a good indication that maybe it should be included right into Trados. And this is exactly what happened here. The SDL XLIFF Converter for MS Office is by far the most popular app, ergo it will now be available right within SDL Trados Studio 2011 by default -- albeit with slightly more advanced features. 

ADVERTISEMENT

memoQ 6.0 is now released

memoQ 6.0 enables real-time preview for Trados TagEditor files, importing Microsoft Word and Excel files without Microsoft Office installed, predictive typing of translation results, integration with Microsoft Translator, Systran Enterprise Server, LetsMT, more versioning, content connectors for local file folders, and... much more!

Attend introductory webinars, download the free 45-day trial from http://kilgray.com and join the ever-growing team of memoQ users!

2. Verifika (Premium Edition)

When I wanted to talk to Julia about Verifika, the QA tool that her company sells, I learned one thing very quickly: it's hard to connect with someone in Siberia when you live on the US West Coast (star-crossed time zones!). Still, after we managed to connect after several attempts, she had a chance to show me a tool that can keep up with the best of them. And you don't have to be a techie to be able to use it.

Here's what Verifika does: it allows you to perform quality assurance on bilingual file formats (presently TMX, Trados TTX, and Trados and memoQ XLIFF files are supported, with Idiom XLZ, SDLX ITD, and Wordfast TXML files in the immediate pipeline). While most translation environment tools have some kind of internal quality assurance feature nowadays, that set of features is naturally limited, a restriction that tools like Verifika don't suffer from. The main categories that the individual QA checks are grouped under include spelling (either with the Word or the Hunspell checker), terminology (which can be added to and altered on the fly), consistency, and then a variety of  subcategories under "Common," including omissions, untranslatables, punctuation, spacing, quotes, measurements, tags, numbers, etcetera, etcetera, most of which are preconfigured for the most common languages.

I really like the tool's ease of use, super-transparent interface, and ability to change errors right from the report the tool generates in the actual translation file (a feature that can be disabled if you don't like it).

When Palex, the LSP behind the tool, first offered Verifika, they did so at a price point that may have reflected their development efforts but not what the market was willing to bear. So they lowered it significantly, starting at 75 euros for a freelance version and 300 for an LSP version.

I'm always thrilled about the wide geographical range our tool developers hail from, and it's a pleasure to welcome someone from Siberia into the fold.  

3. MTV -- Machine Translation Views

I've written about the Microsoft Translator Hub, which officially went live last week after a relatively short beta phase. That brevity may not be too surprising knowing what kind of impact this tool should have on the machine translation market. I anticipate the first and most obvious impact to be that Google will quickly come out with a similar system (in the last newsletter I noted that they've seemed to be playing catch-up with Microsoft in the area of MT lately). But more significantly, I predict that this might have a profound impact on technology vendors who offer trainable rules-based or statistical machine translation systems, and more particularly those that go for the large enterprise accounts.

Why? The Microsoft Translator Hub allows you to feed your TMX files (translation memory exchange files) into the system. On the basis of those -- and on top of the existing MT capabilities of Microsoft/Bing Translator -- a machine translation engine is created that is specific to your terminology and style. Of course, this is what statistical MT vendors like Asia Online and other solutions built around the open-source Moses engine have been offering for some time, with the difference that here you start off with a relatively decent translation engine in the first place that you just refine with your own data. Rules-based MT vendors like Systran or PROMT offer the same principle (an existing engine that you continue to refine), but the refinement is much more labor-intensive (I'm working on one right now, so I REALLY feel that point) and expensive when it comes to enterprise-wide implementations.

So, what's the downside -- aside from another step forward for machine translation developers that some of you might not readily embrace? It might be this clause in the terms of use:

By sharing content on the Service, Microsoft and its affiliates, resellers, distributors, service providers, partners, and suppliers may use, modify, adapt, reproduce, distribute, publish and display it. This includes your name or other information you supply with the content. These rights apply solely to the extent necessary to provide the Service.

It's already been discussed to the n-th degree with Google's translation services, but the data that you provide will be used by Microsoft to further strengthen its own MT engine and possibly other services.

It'll be interesting to see the reaction of Lionbridge's customers to that clause. Lionbridge announced this week that it has dropped IBM's MT engine (after it announced that partnership to great fanfare just less than two years ago) and is now working with the Microsoft Translator Hub. (As a side note: It boggles my mind how often companies like Lionbridge change direction mid-course. If I did the same, you would not be reading this 211th edition of the newsletter -- I would have stopped after the 15th edition or so.)

 

Another news item in the world of machine translation promises to be interesting for us as well. Awhile back I profiled Multilizer, the company that had been developing and marketing a localization tool for a long time but had seen the writing on the wall and chosen to orient itself in other directions as well. I mentioned that the company was working on a product that would allow you to automatically compare different machine translations of the same text and -- segment-by-segment -- come up with the best translation.

The tool has finally been released, and it's called MT-Qualifier. I talked this morning to Niko Papula, Multilizer's CEO, and though there was much he could not reveal (his lawyer has given him a strict gag order regarding the underlying technology because of a pending patent), there was definitely enough to make it worth a mention.

Niko's company is from Finland, and I realized again that I really enjoy talking to Northern European business people. They tend to eschew fluff und hyperbole, remaining resolutely down-to-earth when it comes to talking about what they do and how they do it. According to Niko, MT-Qualifier currently works in only one language combination -- English<>Spanish. When will there be other language combinations? Well, says Niko, let's see. We'll try getting it right with this single one and then see what other demand is out there. Can you imagine that kind of response from a representative of a US company or one that is trying to act like one?

The other limitation of the comparison engine is that you can't use customized MT engines in the process. You can have as many engines as you want, but all of them must be generic, without any specialized lexicons or customized data. Pricing is still up in the air, but right now they're thinking about approximately 1 euro per page (whatever "page" means exactly). This will not only include the qualifying of the different outputs but the usage of the various machine translation programs as well.

I'm eager to see where this idea and product will go.

 

There was a very enjoyable blog posting by the daughter of a machine translation developer that might resonate with some of you. In it she attempts to process her dad's loss of a job. You can find more right here. Also interesting was this article that is pondering the copyright limitations on machine translation.

Links to these kinds of articles and much, much more can be accessed in my Twitter stream -- I encourage you to subscribe.

 

And last but not least in this section, the Association of Machine Translation in the Americas (AMTA -- I've already made enough jokes about this unwieldy name, so I won't repeat those now) is having its conference right after the ATA conference in San Diego on October 28 through November 1. (The ATA is from October 24 through the 27 and the site of the official launch of Found in Translation!) I encourage you to consider dropping in at the AMTA conference this year. Last month I mentioned that Duolingo founder Luis von Ahn will be giving one of the keynotes, which I find very exciting. I will also give some tutorials about using machine translation in translation environment tools (I know, I know -- not quite the same caliber, but, hey, I'm trying.)

ADVERTISEMENT

New and available: LSP-focused offers for Across Language Server

Do you want to extend your business with a market-leading translation management software to reach new customers?

Across offers special-priced, new bundles focused on the needs of small, medium and large Language Service Providers.

To facilitate your entry or migration from other systems, Across provides installation support and offers individual training for you.

Request your individual offer now!

4. Excelling Glossary Searches

Stanislav Okhvat must really love us. He is our industry's Energizer bunny, producing one tool after the other and . . .  just giving them away!

A few months ago I talked about his plug-ins for Word, Excel, Visio, and AutoCAD that perform a variety of functions to prepare files for a TEnT-based translation process or to post-process them after translation.

Now he has a number of other tools as well. One, the Translation Tool Utility, allows you to paste unformatted text or special characters and strip tags out of segments in memoQ (anyone want to guess which TEnT Stanislav himself uses?).

Another is the Glossary Manager, which is part of the Excel plug-in. I had read about the glossary tool some time ago in an email from Stanislav, though I hadn't taken it too seriously. But now I've had a chance to look at it and I think it's very clever. It may not quite reach the power of APSiC Xbench -- after all, it supports only glossaries in Excel -- but it's great if most of your glossaries are in Excel and you would like to find a way to quickly search from within any Windows application by simply highlighting a term and pressing a certain keyboard combination. You can even search any number of Excel glossaries all at once, the setup takes all of two minutes (he has decent instructions on his website), and it is extremely customizable.

The installer for his tools includes all the tools he offers, but you can choose to install only the applications that you really want in step 2 or 3 of the installation process.

This is not a tool that will finally make the one big difference in your life you've been awaiting for years, but I guarantee that you'll like it if you need a good solution for searching your Excel files and you're sick of slowly and manually going through them. 

5. The Panic-Stricken Translator (Premium Edition)

. . . would be a good appellation for manysome quite a few of us. Every few weeks when there is another announcement that promises an industry-changing technology or product, I receive a good number of emails from colleagues asking me what I make of this and whether this could be the deathblow to our profession that we all had feared all along.

The answer is always: No.

Our industry is changing. No doubt.

  • The size of our markets has dramatically changed -- both in a good sense: we can and do work for clients around the world -- and in a bad way -- competition from everywhere has also risen dramatically.
  • The size of some clients, the complexity of some projects, and the natural market forces of a maturing industry have led to the growth and formation of some very large language service providers -- with all their benefits and drawbacks.
  • Technology has become a very integral part of our everyday work lives, making it more and more unlikely to "opt out" (though I have a strong feeling that those who have chosen not to use any technology beyond a word processor and email and are making a good living so far will still be able to do this for a long time to come).
  • The ever-growing presence of machine translation has led to a different set of expectations for many customers.
  • Workflow models like crowdsourcing have come out into the open (they've been around for a long time!), accompanied by the participation of untrained translators in certain translation projects.
  • Portals like ProZ have furthered a segmentation of our industry into lower-wage and higher-wage workers.

And I could go on and on (or ask any of you to continue this list).

But here is what I know: my wife and kids are still frustrated about me working too much, I have been able to slowly raise my rates, and some of the projects I do are really boring and dull while others are a lot of fun -- all pretty much the same as it was 10 or 15 years ago.

Do you remember when Google announced Google Translator Toolkit? Or when Lionbridge announced its Translation Workspace product? The world seemed to come to a halt. Well, not quite, but many of us assumed that these products would have a tremendous impact on our industry. (And I might be as guilty as anyone there.)  The truth is that these products really had no or very little impact. If you work for Lionbridge you might be using the Workspace product; otherwise it's very unlikely, despite Lionbridge's many efforts to make this a game changer. And Google's Toolkit product? To their credit, they never really marketed it to professional translators -- and that's good, because I think that very, very few are using it.

Big companies or companies that think of themselves as big make big announcements. Many times all that is left is just that: the announcement.

And while things have changed for us to some degree, overall we're still doing essentially the same job we did 10 or 20 years ago. And the same that those who come after us will do as well (see the introduction). 

ADVERTISEMENT

The Natural Choice for Translators

Find out, this month, why we think SDL Trados Studio 2011 is the natural choice for translators worldwide. You could also be in with a chance of adopting an endangered animal...

Learn more »

We are also offering all Toolbox readers the opportunity to save 30% off the leading translation memory software -- SDL Trados Studio 2011. Furthermore, upgrade from SDL Trados 2007 and save up to 40% off before 31st July.

Save today »

6. New Password for the Tool Kit Archive

As a subscriber to the Premium version of this newsletter you have access to an archive of Premium newsletters going back to May 2008.

You can access the archive right here. This month the user name is toolbox and the password is foundintranslation.

New user names and passwords will be announced in future newsletters.

The Last Word on the Tool Box Newsletter

If you would like to promote this newsletter by placing a link on your website, I will in turn mention your website in a future edition of the Tool Box newsletter. Just paste the code you find here into the HTML code of your webpage, and the little icon that is displayed on that page with a link to my website will be displayed.

Here is a website that added the link:

www.jlperello.com 

© 2012 International Writers' Group